U.S. patent application number 14/532838 was filed with the patent office on 2016-03-03 for image sensor having depth detection pixels and method for generating depth data with the image sensor.
The applicant listed for this patent is SK hynix Inc.. Invention is credited to Chang-Hee PYEOUN.
Application Number | 20160065870 14/532838 |
Document ID | / |
Family ID | 55404059 |
Filed Date | 2016-03-03 |
United States Patent
Application |
20160065870 |
Kind Code |
A1 |
PYEOUN; Chang-Hee |
March 3, 2016 |
IMAGE SENSOR HAVING DEPTH DETECTION PIXELS AND METHOD FOR
GENERATING DEPTH DATA WITH THE IMAGE SENSOR
Abstract
An image sensor includes: a plurality of pixels that include a
plurality of color detection pixels and a plurality of depth
detection pixels, wherein the plurality of pixels are arranged in
two dimensions; and a photoelectric conversion layer formed to
correspond to the plurality of the pixels, wherein each of the
plurality of the color detection pixels comprises: a first light
condensing layer disposed over the photoelectric conversion layer;
and a band pass filter layer interposed between the photoelectric
conversion layer and the first light condensing layer, and wherein
each of the plurality of the depth detection pixels comprises: a
second light condensing layer disposed over the photoelectric
conversion layer and having a greater size than a size of the first
light condensing layer.
Inventors: |
PYEOUN; Chang-Hee;
(Gyeonggi-do, KR) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
SK hynix Inc. |
Gyeonggi-do |
|
KR |
|
|
Family ID: |
55404059 |
Appl. No.: |
14/532838 |
Filed: |
November 4, 2014 |
Current U.S.
Class: |
348/302 |
Current CPC
Class: |
H01L 27/14649 20130101;
H01L 27/14621 20130101; H04N 9/04557 20180801; G02B 5/201 20130101;
H01L 27/14629 20130101; H04N 5/3696 20130101; H01L 27/14645
20130101; H04N 5/36961 20180801; H04N 9/045 20130101; H04N 5/36965
20180801; H04N 5/369 20130101; H01L 27/14623 20130101; H01L
27/14627 20130101 |
International
Class: |
H04N 5/369 20060101
H04N005/369; H04N 5/225 20060101 H04N005/225; H04N 5/378 20060101
H04N005/378; H04N 9/04 20060101 H04N009/04; G02B 5/20 20060101
G02B005/20 |
Foreign Application Data
Date |
Code |
Application Number |
Aug 28, 2014 |
KR |
10-2014-0112887 |
Claims
1. An image sensor, comprising: a plurality of pixels that include
a plurality of color detection pixels and a plurality of depth
detection pixels, wherein the plurality of pixels are arranged in
two dimensions; and a photoelectric conversion layer formed to
correspond to the plurality of the pixels, wherein each of the
plurality of the color detection pixels comprises: a first light
condensing layer disposed over the photoelectric conversion layer;
and a band pass filter layer interposed between the photoelectric
conversion layer and the first light condensing layer, and wherein
each of the plurality of the depth detection pixels comprises: a
second light condensing layer disposed over the photoelectric
conversion layer and having a greater size than a size of the first
light condensing layer.
2. The image sensor of claim 1, further comprising: a dielectric
layer formed over the photoelectric conversion layer, having a
planar upper surface, and including the band pass filter layer,
wherein the first light condensing layer and the second light
condensing layer are formed over the dielectric layer.
3. The image sensor of claim 1, wherein the band pass filter layer
extends to the plurality of depth detection pixels to be interposed
between the photoelectric conversion layer and the second light
condensing layer.
4. The image sensor of claim 3, wherein the band pass filter layer
of the depth detection pixels is thicker than the band pass filter
layer of the color detection pixels.
5. The image sensor of claim 1, wherein two or more of the depth
detection pixels share one second light condensing layer.
6. The image sensor of claim 1, wherein a thickness of the first
light condensing layer is the same as a thickness of the second
light condensing layer, or the first light condensing layer is
thinner than the second light condensing layer.
7. The image sensor of claim 1, wherein each of the first light
condensing layer and the second light condensing layer includes a
stacked structure of a plurality of material layers, wherein an
upper layer of the first light condensing layer has a smaller size
than a lower layer of the first light condensing layer, and wherein
an upper layer of the second light condensing layer has a smaller
size than a lower layer of the second light condensing layer.
8. The image sensor of claim 7, wherein a size of a lowermost layer
of the first light condensing layer is shorter than a size of a
lowermost layer of the second light condensing layer.
9. The image sensor of claim 7, wherein a total thickness of the
first light condensing layer is the same as a total thickness of
the second light condensing layer, and wherein a lowermost layer of
the second light condensing layer is thinner than a lowermost layer
of the first light condensing layer.
10. The image sensor of claim 1, wherein the first light condensing
layer includes a hemispheric lens, wherein the second light
condensing layer includes a stacked structure of a plurality of
material layers, and wherein an upper layer of the plurality of
material layers has a smaller size than a lower layer of the
plurality of material layers.
11. The image sensor of claim 1, wherein the band pass filter layer
includes a stacked structure of a plurality of filter layers, and
wherein the plurality of filter layers have different refraction
indices from each other.
12. The image sensor of claim 1, wherein the plurality of color
detection pixels includes first color pixels, second color pixels,
and third color pixels, wherein the band pass filter layer of the
first color pixels is thicker than the band pass filter layer of
the second color pixels, and wherein the band pass filter layer of
the second color pixels is thicker than the band pass filter layer
of the third color pixels.
13. The image sensor of claim 12, wherein the first color pixels
include red pixels, and the second color pixels include green
pixels, and the third color pixels include blue pixels.
14. The image sensor of claim 1, wherein the plurality of pixels
further comprise a plurality of phase difference detection pixels,
and wherein each of the plurality of the phase difference detection
pixels comprises: a third light condensing layer disposed over the
photoelectric conversion layer; and a light blocking layer
interposed between the photoelectric conversion layer and the third
light condensing layer and having an opening that is eccentrically
disposed with respect to an optical axis of the photoelectric
conversion layer.
15. The image sensor of claim 14, wherein the third light
condensing layer includes a hemispheric lens or a stacked structure
of a plurality of material layers, and wherein an upper layer of
the plurality of material layers has a smaller size than a lower
layer of the plurality of material layers.
16. A method for generating depth data in an image sensor provided
with a plurality of pixels including a plurality of color detection
pixels, a plurality of depth detection pixels, and a plurality of
phase difference detection pixels that are arranged in two
dimensions, comprising: calculating a blur weight value from the
plurality of phase difference detection pixels; extracting a first
edge value from the plurality of color detection pixels; extracting
a second edge value from the plurality of depth detection pixels;
calculating a first blur value and a second blur value based on the
first edge value and the second edge value using a Point Spread
Function; amplifying the first blur value and the second blur value
by reflecting the blur weight value into the first blur value and
the second blur value; and generating depth information based on a
difference between the amplified first blur value and the amplified
second blur value.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] The present application claims priority of Korean Patent
Application No. 10-2014-0112887, filed on Aug. 28, 2014, which is
incorporated herein by reference in its entirety.
BACKGROUND
[0002] 1. Field
[0003] Exemplary embodiments relate to a semiconductor fabrication
technology and, more particularly, to an image sensor having depth
detection pixels, and a method for generating depth data by using
the image sensor.
[0004] 2. Description of the Related Art
[0005] Image sensors transform optical signals including image or
depth data into electrical signals. Researchers and the industry
are ardently studying and developing high-precision
three-dimensional (3D) image sensors that provide depth data in
addition to image data.
SUMMARY
[0006] An embodiment is directed to an image sensor including depth
detection pixels, and a method for generating depth data by using
the image sensor.
[0007] In accordance with an embodiment, an image sensor includes:
a plurality of pixels that include a plurality of color detection
pixels and a plurality of depth detection pixels, wherein the
plurality of pixels are arranged in two dimensions; and a
photoelectric conversion layer formed to correspond to the
plurality of the pixels, wherein each of the plurality of the color
detection pixels comprises: a first light condensing layer disposed
over the photoelectric conversion layer, and a band pass filter
layer interposed between the photoelectric conversion layer and the
first light condensing layer, and wherein each of the plurality of
the depth detection pixels comprises: a second light condensing
layer disposed over the photoelectric conversion layer and having a
greater size than a size of the first light condensing layer.
[0008] The image sensor may further include: a dielectric layer
formed over the photoelectric conversion layer, having a planar
upper surface, and including the band pass filter layer, wherein
the first light condensing layer and the second light condensing
layer are formed over the dielectric layer.
[0009] The band pass filter layer may extend to the plurality of
depth detection pixels to be interposed between the photoelectric
conversion layer and the second light condensing layer. The band
pass filter layer of the depth detection pixels may be thicker than
the band pass filter layer of the color detection pixels.
[0010] Two or more of the depth detection pixels may share one
second light condensing layer. The thickness of the first light
condensing layer may be the same as the thickness of the second
light condensing layer, or the first light condensing layer may be
thinner than the second light condensing layer.
[0011] Each of the first light condensing layer and the second
light condensing layer includes a stacked structure of a plurality
of material layers, wherein an upper layer of the first light
condensing layer has a smaller size than a lower layer of the first
light condensing layer, and wherein an upper layer of the second
light condensing layer has a smaller size than a lower layer of the
second light condensing layer. The size of a lowermost layer of the
first light condensing layer may be shorter than a size of a
lowermost layer of the second light condensing layer. The total
thickness of the first light condensing layer is the same as a
total thickness of the second light condensing layer, and wherein a
lowermost layer of the second light condensing layer is thinner
than a lowermost layer of the first light condensing layer.
[0012] The first light condensing layer may includes a hemispheric
lens, wherein the second light condensing layer may includes a
stacked structure of a plurality of material layers, and wherein an
upper layer of the plurality of material layers has a smaller size
than a lower layer of the plurality of material layers.
[0013] The band pass filter layer may include a stacked structure
of a plurality of filter layers, and wherein the plurality of
filter layers has different refraction indices from each other.
[0014] The plurality of the color detection pixels may includes
first color pixels, second color pixels, and third color pixels,
wherein the band pass filter layer of the first color pixels is
thicker than the band pass filter layer of the second color pixels,
and wherein the band pass filter layer of the second color pixels
is thicker than the band pass filter layer of the third color
pixels.
[0015] The first color pixels may include red pixels, and the
second color pixels may include green pixels, and the third color
pixels may include blue pixels.
[0016] The plurality of the pixels may further include a plurality
of phase difference detection pixels, and wherein each of the
plurality of the phase difference detection pixels may include: a
third light condensing layer disposed over the photoelectric
conversion layer; and a light blocking layer interposed between the
photoelectric conversion layer and the third light condensing layer
and having an opening that is eccentrically disposed based on with
respect to an optical axis of the photoelectric conversion
layer.
[0017] The third light condensing layer includes a hemispheric lens
or a stacked structure of a plurality of material layers, and
wherein an upper layer of the plurality of material layers has a
smaller size than a lower layer of the plurality of material
layers.
[0018] In accordance with another embodiment, a method for
generating depth data in an image sensor provided with a plurality
of pixels including a plurality of color detection pixels, a
plurality of depth detection pixels, and a plurality of phase
difference detection pixels that are arranged in two dimensions,
comprising: calculating a blur weight value from the plurality of
phase difference detection pixels; extracting a first edge value
from the plurality of color detection pixels; extracting a second
edge value from the plurality of depth detection pixels;
calculating a first blur value and a second blur value based on the
first edge value and the second edge value using a Point Spread
Function; amplifying the first blur value and the second blur value
by reflecting the blur weight value into the first blur value and
the second blur value; and generating depth information based on a
difference between the amplified first blur value and the amplified
second blur value.
BRIEF DESCRIPTION OF THE DRAWINGS
[0019] FIG. 1 is a block view illustrating an image sensor
accordance with an embodiment.
[0020] FIGS. 2A to 2C are plan views illustrating a pixel array of
the image sensor in accordance with embodiments.
[0021] FIGS. 3A and 3B illustrate an image sensor in accordance
with an embodiment.
[0022] FIGS. 4A and 4B illustrate an image sensor in accordance
with an embodiment.
[0023] FIG. 5 is a cross-sectional view illustrating a phase
difference detection pixel of the image sensor in accordance with
an embodiment.
[0024] FIG. 6 is a block view illustrating an Image Signal
Processor (ISP) of the image sensor in accordance an
embodiment.
[0025] FIG. 7 is a flowchart describing a method for generating
depth data in an image sensor in accordance with an embodiment.
DETAILED DESCRIPTION
[0026] Exemplary embodiments will be described below in more detail
with reference to the accompanying drawings. The embodiments may,
however, be modified in different forms and should not be construed
as limited to the configurations set forth herein. Throughout the
disclosure, like reference numerals refer to like parts.
[0027] The drawings are not necessarily to scale and, in some
instances proportions may have been exaggerated in order to clearly
illustrate features of the embodiments. When a first layer is
referred to as being "on" a second layer or "on" a substrate, it
not only refers to where the first layer is formed directly on the
second layer or the substrate but also where a third layer exists
between the first layer and the second layer or the substrate.
[0028] The following embodiments provide an image sensor including
depth detection pixels and a method for generating depth data by
using the image sensor. Depth data is essential to realizing a
three-dimensional (3D) image. Diverse technologies, which have been
suggested to generate depth data, include Time Of Flight (TOF),
triangulation, an infrared ray (IR) radiation pattern detection
method, semi-automatic algorithm and so forth. The methods require
a light source, such as an infrared ray light source to make a
depth difference. However, it has a problem in that depth data is
not reliably produced in high illumination environments (such as
outdoor environment in daytime), due to interferences. Also, the
triangulation technique and a stereo camera technique may generate
highly accurate depth data, but their structures are complicated
and their prices are expensive because they require a plurality of
cameras and expensive equipment.
[0029] To address these issues, the following embodiments provide
an image sensor including a plurality of color detection pixels and
a plurality of depth detection pixels. The depth detection pixels
have a different focal length from that of the color detection
pixels in a pixel array. In the pixel array, a plurality of pixels
is arranged in two dimensions (2D). The focal length may be a focal
length between an object and each pixel, and the focal length from
the depth detection pixels may be longer than the focal length from
the color detection pixels. Under a structure where the color
detection pixels and the depth detection pixels are integrated in
the pixel array and have different focal lengths from each other,
interference from external environment, e.g., a high illumination
environment, may be effectively suppressed. As a result, accurate
depth data may be generated. Also, since the image sensor has a
simple structure and is suitable for mass production, the image
sensor may be provided at an economical price.
[0030] FIG. 1 is a block view illustrating an image sensor in
accordance with an embodiment.
[0031] Referring to FIG. 1, the image sensor 1000 may include a
pixel array 1100, an Image Signal Processor (ISP) 1110, a row
driver 1120, a Correlated Double Sampler (CDS) 1130, an
Analog-Digital Converter (ADC) 1140, a ramp signal generator 1160,
a timing generator 1170, and a control register 1180.
[0032] The pixel array 1100 may include a plurality of pixels for
generating pixel signals based on the intensity of incident light.
Each of the pixels may include at least one pixel transistor and a
photoelectric conversion element. Each pixel transistor may include
a transfer transistor, a reset transistor, a drive transistor, and
a select transistor. The transfer transistor is coupled with the
photoelectric conversion element. The reset transistor shares a
floating diffusion region with the transfer transistor. The drive
transistor has its gate coupled with the floating diffusion region.
The select transistor is coupled in series with the drive
transistor. The photoelectric conversion element may include a
photodiode or a pinned photodiode. The pixels of the pixel array
1100 may include a plurality of color detection pixels, a plurality
of depth detection pixels, and a plurality of phase difference
detection pixels, which will be described in detail below.
[0033] The timing generator 1170 may apply a control signal or a
dock signal to the row driver 1120, the correlated double sampler
1130, the analog-digital converter 1140, and the ramp signal
generator 1160 to control given operations or timings. The control
register 1180 may store various commands necessary for operation of
the image sensor 1000.
[0034] The row driver 1120 may decode the control signal or the
clock signal that is outputted from the timing generator 1170 and
supply a gate signal to each row of the pixel array 1100. The pixel
array 1100 may output a pixel signal. The pixel signal is outputted
from a row which is selected based on the gate signal transmitted
from the row driver 1120. In other words, the digital pixel signal
(DPS) transmitted from the pixel array 1100 to the image signal'
processor 1110 may be provided through a certain selected row.
[0035] The image signal processor 1110 processes the digital pixel
signal (DPS) outputted from the pixel array 1100 through the
correlated double sampler 1130 and the analog-digital converter
1140. The digital pixel signal (DPS) is outputted in form of image
data that human beings may see. The image signal processor 1110 may
include a control block, a memory block, a phase difference
detection block including a blur weight value generation unit, an
edge detection block, and a depth detection block, which will be
described below.
[0036] FIGS. 2A to 2C are plan views illustrating pixel arrays of
an image sensor in accordance with an embodiment. FIG. 2A is a plan
view illustrating a pixel array of an image sensor in accordance
with an embodiment, and FIGS. 2B and 2C are plan views illustrating
pixel arrays of the image sensor in accordance with other
embodiments.
[0037] Referring to FIG. 2A, the pixel array 1100 of the image
sensor 1000 may include a plurality of pixels that are arranged in
two dimensions, and the pixels may include a plurality of color
detection pixels 110, a plurality of depth detection pixels 120,
and a plurality of phase difference detection pixels 130. The color
detection pixels 110 may be arranged based on a Bayer arrangement
and the depth detection pixels 120 and the phase difference
detection pixels 130 may be randomly disposed in such a manner that
the depth detection pixels 120 and the phase difference detection
pixels 130 replace the color detection pixels 110 at particular
locations. The phase difference detection pixels 130 may be
randomly disposed by pixel group unit. In an embodiment, one pixel
group may include two pixels 131 and 132 and may include light
blocking layers 251, 252. See FIGS. 2A-2C and 5. The light blocking
layers 251, 252 have openings 251A, 252A that are eccentrically
disposed with respect, to a center axis of the phase difference
detection pixels 130.
[0038] In the pixel array 1100, a plurality of color detection
pixels 110 and one or more depth detection pixels 120 may be
disposed in every single row and/or column. One or more phase
difference detection pixels 130 may be disposed in every single row
and/or column, or in one or more rows and/or columns.
[0039] Also, in the pixel array 1100, a plurality of color
detection pixels 110 may be disposed in every single row and/or
column, and one or more depth detection pixels 120 may be disposed
in one or more rows and/or columns. Two or more phase difference
detection pixels 130 may be disposed in the same row and/or column
where the depth detection pixels 120 are disposed, or two or more
phase difference detection pixels 130 may be disposed in a row
and/or a column different from one where the depth detection pixels
120 are disposed.
[0040] The color detection pixels 110 may include first color
pixels 111, second color pixels 112, and third color pixels 113.
The first to third color pixels 111 to 113 may be arranged in a
Bayer pattern. The first color pixels 111 may be red pixels R, and
the second color pixels 112 may be green pixels G, and the third
color pixels 113 may be blue pixels B. Alternatively, the first to
third color pixels 111 to 113 may be Cyan pixels, Magenta pixels,
and Yellow pixels, respectively.
[0041] The depth detection pixels 120 may be disposed randomly by
replacing one of the color detection pixels 110 that are arranged
in the Bayer pattern. For example, the depth detection pixels 120
may replace green pixels G in a 2.times.2 array. See reference
numeral {circle around (1)} in FIG. 2A. Also, in a 2.times.4 array,
the depth detection pixels 120 may replace all the color detection
pixels 110 corresponding to a 2.times.2 array. See the reference
numeral {circle around (2)} in FIG. 2A. As a result, the color
detection pixels 110 are in a 2.times.2 array.
[0042] Subsequently, referring to FIG. 2B, the color detection
pixels 110, the depth detection pixels 120, and the phase
difference detection pixels 130 may be disposed row by row. In
other words, the pixel array 1100 may include a first row where the
color detection pixels 110 are disposed, a second row where the
depth detection pixels 120 are disposed, and a third row where the
phase difference detection pixels 130 are disposed. The first to
third rows may be adjacent to each other in a vertical direction,
or they may be spaced apart from each other and randomly located.
In another embodiment, the color detection pixels 110, the depth
detection pixels 120, and the phase difference detection pixels 130
may be disposed column by column.
[0043] Subsequently, referring to FIG. 2C the color detection
pixels 110, the depth detection pixels 120, and the phase
difference detection pixels 130 may be disposed block by block (or
region by region). The number of the blocks where the pixels are
disposed may be one or more. The area and the shape of the blocks
where the pixels are disposed may vary according to given
specifications and desired characteristics.
[0044] FIGS. 3A and 36 illustrate an image sensor in accordance
with a first embodiment. FIG. 3A is a plan view illustrating a
portion of a pixel array of the image sensor in accordance with the
first embodiment, and FIG. 3B is a cross-sectional view of color
detection pixels and depth detection pixels that is taken along an
A-A' line shown in FIG. 3A. FIGS. 4A and 4B illustrate an image
sensor in accordance with a second embodiment. FIG. 4A is a plan
view illustrating a portion of a pixel array of the image sensor in
accordance with the second embodiment, and FIG. 4B is a
cross-sectional view of color detection pixels and depth detection
pixels that is taken along an A-A' line shown in FIG. 4A. FIG. 5 is
a cross-sectional view illustrating phase difference detection
pixels of the image sensors in accordance with the first and second
embodiments along a B-B' line shown in FIGS. 3A and 4A.
[0045] Referring to FIGS. 3A, 3B, 4A and 4B, each of the image
sensors in accordance with the first and second embodiments may
include a pixel array and a photoelectric conversion layer 202. In
each pixel array, a plurality of pixels including a plurality of
color detection pixels 110 and a plurality of depth detection
pixels 120 are arranged in two dimensions.
[0046] The photoelectric conversion layers 202 corresponding to the
color detection pixels 110 and the depth detection pixels 120 may
have the same light receiving portions, in area and in depth. In
another embodiment, the photoelectric conversion layers 202
corresponding to the color detection pixels 110 and the depth
detection pixels 120 may have the same light receiving portions in
terms of area, but may be different in terms of depth. For example,
the depth of the photoelectric conversion layer 202 corresponding
to the depth detection pixels 120 may be deeper than the depth of
the photoelectric conversion layer 202 corresponding to the color
detection pixels 110. In this case, quantum efficiency of the depth
detection pixels 120 and a focal length to the object may be
effectively enhanced.
[0047] The photoelectric conversion layer 202 may be formed in a
substrate 201. The photoelectric conversion layer 202 formed in the
substrate 201 may include a plurality of photoelectric converters
that vertically overlap with one another. Each of the photoelectric
converters may be a photodiode including an N-type impurity region
and a P-type impurity region. The substrate 201 may be
semiconductor substrate, and it may include a monocrystalline
silicon-containing material.
[0048] The color detection pixels 110 in the image sensor in
accordance with the first and second embodiments may include first
color pixels 111, second color pixels 112, and third color pixels
113. The first color pixels 111 may be red pixels R, and the second
color pixels 112 may be green pixels G. The third color pixels 113
may be blue pixels B.
[0049] The color detection pixels 110, which are the first color
pixels 111 to the third color pixels 113, may include the
photoelectric conversion layer 202, a band pass filter layer 210
disposed over the photoelectric conversion layer 202, a dielectric
layer 205 disposed over the band pass filter layer 210, and a first
light condensing layer 220 disposed over the dielectric layer 205.
The dielectric layer 205 is formed to cover the band pass filter
layer 210 over a contour of the substrate 201 including the
photoelectric conversion layer 202. The dielectric layer 205 may
have a planar upper surface to improve characteristics of the first
light condensing layer 220.
[0050] In the color detection pixels 110, incident light with a
particular wavelength band passes through the band pass filter
layer 210. In short, the band pass filter layer 210 substantially
performs the function of a color filter. To this end, the
thicknesses of the band pass filter layer 210 corresponding to the
first color pixels 111, the second color pixels 112, and the third
color pixels 113 may be different from each other. To be specific,
the thickness of a first band pass filter layer 211 formed over the
first color pixels 111 may be thicker than the thickness of a
second band pass filter layer 212 formed over the second color
pixels 112. The thickness of the second band pass filter layer 212
formed over the second color pixels 112 may be thicker than the
thickness of a third band pass filter layer 213 formed over the
third color pixels 113. Since the band pass filter layer 210 is
disposed between the photoelectric conversion layer 202 and the
first light condensing layer 220, optical crosstalk between the
neighboring pixels may be effectively prevented. The band pass
filter layer 210 may be a stacked structure where a plurality of
filter layers, having different refraction indices, is alternately
stacked. The filter layers may include an oxide, a nitride, or an
oxynitride. For example, the band pass filter layer 210 may be a
stacked structure where a first filter layer 203 and a second
filter layer 204 are alternately stacked, and the first filter
layer 203 may include an aluminum nitride or a silicon oxide, and
the second filter layer 204 may include a silicon nitride or a
titanium oxide.
[0051] The first light condensing layer 220 may be a hemispheric
lens or a stacked structure of a plurality of material layers.
Upper layers of the plurality of material layers have a shorter
diameter than the lower layers of the plurality of material layers.
The material layers may include an oxide, a nitride, or an
oxynitride. As illustrated in the drawings, when the first light
condensing layer 220 is a stacked structure, the stacked structure
may include a first layer 221 and a second layer 222. The second
layer 222 is formed over the first layer 221 and has a shorter
diameter than the first layer 221. The first layer 221 and the
second layer 222 may be formed of the same material, or different
materials. For example, the first layer 221 may include a silicon
nitride, while the second layer 222 may include a silicon oxide.
Also, the thickness t1 of the first layer 221 may be the same as
the thickness t2 of the second layer 222 (t1=t2), or the thickness
t1 of the first layer 221 may be different from the thickness t2 of
the second layer 222 (t1.noteq.t2). The focal length to the object
may be adjusted by controlling the thickness t1 of the first layer
221 and the thickness t2 of the second layer 222 while the
thickness T1 (T1=t1+t2) of the first light condensing layer 220 is
fixed. For example, while the thickness T1 of the first light
condensing layer 220 is fixed, the thickness t1 of the first layer
221 and the thickness t2 of the second layer 222 may be inversely
proportional to each other.
[0052] Therefore, as the thickness t1 of the first layer 221
increases, the focal length may become shorter. Also, as the
thickness t1 of the first layer 221 decreases, the focal length may
become longer.
[0053] The depth detection pixels 120 in the image sensor in
accordance with the first and second embodiments may include the
photoelectric conversion layer 202, the dielectric layer 205
disposed over the photoelectric conversion layer 202, and a second
light condensing layer 230 disposed over the dielectric layer 205.
The second light condensing layer 230 may be formed per the depth
detection pixels 120 (refer to FIGS. 3A and 3B), or two or more
depth detection pixels 120 may share one second light condensing
layer 230 (refer to FIGS. 4A and 4B).
[0054] The photoelectric conversion layer 202 corresponding to the
depth detection pixels 120 may receive an infrared ray. Therefore,
when a depth of the photoelectric conversion layer 202
corresponding to the depth detection pixels 120 is deeper than a
depth of the photoelectric conversion layer 202 corresponding to
the color detection pixels 110, quantum efficiency and a focal
length of the depth detection pixels 120 may be effectively
enhanced. This is because the infrared ray is absorbed deeper than
the visible light in the photoelectric conversion layer 202 that is
formed in the substrate 201 including a monocrystalline
silicon-containing material.
[0055] Also, the depth detection pixels 120 may further include the
band pass filter layer 210 that is formed over the depth detection
pixels 120 and the color detection pixels 110. The band pass filter
layer 210 is interposed between the photoelectric conversion layer
202 and the second light condensing layer 230 (refer to FIGS. 3A
and 38). To be specific, the band pass filter layer 210 formed over
the depth detection pixels 120 is denoted as a fourth band pass
filter layer 214. The fourth band pass filter layer 214 performs a
function of cutting off the visible light so that only the infrared
rays are received. In this way, intensity of the infrared rays
entering the depth detection pixels 120 increases, thereby
improving sensor characteristics. To this end, the thickness of the
fourth band pass filter layer 214 over the depth detection pixels
120 may be thicker than the thickness of any of the first band pass
filter layer 211 to the third band pass filter layer 213, which are
portions of the band pass filter layer 210 formed over the color
detection pixels 110.
[0056] The second light condensing layer 230 may be a stacked
structure where a plurality of material layers is stacked. An upper
layer of the plurality of material layers has a shorter diameter
(or a smaller size) than a lower layer. The first layer 221 has a
shorter diameter than the third layer 231 and the second layer 222
has a shorter diameter than the fourth layer 232.
[0057] To be specific, the second light condensing layer 230 may
include a third layer 231 and a fourth layer 232 that is formed
over the third layer 231 and has a shorter diameter than the third
layer 231. The third layer 231 and the fourth layer 232 may be
formed of the same material or different materials. For example,
the third layer 231 may include a silicon nitride, while the fourth
layer 232 may include a silicon oxide. When the first light
condensing layer 220 is a stacked structure, where a plurality of
material layers are stacked, the third layer 231 and the fourth
layer 232 of the second light condensing layer 230 may be formed of
the same material as those of the first layer 221 and the second
layer 222 of the first light condensing layer 220,
respectively.
[0058] The thickness t3 of the third layer 231 may be the same as
thickness t4 of the fourth layer 232 (t3=t4), or the thickness t3
of the third layer 231 may be different from the thickness t4 of
the fourth layer 232 (t3.noteq.t4). The focal length to the object
may be adjusted by manipulating the thickness t3 of the third layer
231 and the thickness t4 of the fourth layer 232 while the
thickness T2 (T2=t3+t4) of the second light condensing layer 230 is
fixed. For example, while the thickness T2 of the second light
condensing layer 230 is fixed, the thickness t3 of the third layer
231 and the thickness t4 of the fourth layer 232 may be inversely
proportional to each other.
[0059] Therefore, as the thickness t3 of the third layer 231 is
increased, the focal length may become shorter. Also, as the
thickness t3 of the third layer 231 is decreased, the focal length
may become longer.
[0060] To provide depth detection pixels 120 which have a different
focal length from the focal length of the color detection pixels
110, the second light condensing layer 230 may have a shape that is
different from that of the first light condensing layer 220. To be
specific, as the focal length between the depth detection pixels
120 and the object is longer than the focal length between the
color detection pixels 110 and the object, depth data is generated
more accurately. To this end, the second light condensing layer 230
may have a greater diameter or area than the first light condensing
layer 220. In terms of the area, when a light receiving area of the
photoelectric conversion layer 202 of the color detection pixels
110 is the same as a light receiving area of the photoelectric
conversion layer 202 of the depth detection pixels 120, the area of
the second light condensing layer 230 corresponding to the
photoelectric conversion layer 202 may be greater than the area of
the first light condensing layer 220 corresponding to the
photoelectric conversion layer 202. Also, the second light
condensing layer 230 may have the same thickness as that of the
first light condensing layer 220 (T1=T2), or the second light
condensing layer 230 may be thicker than the first light condensing
layer 220 (T1<T2).
[0061] When the first light condensing layer 220 and the second
light condensing layer 230 are all stacked structures where a
plurality of material layers are stacked, the diameter or area of
the lowermost layer of the first light condensing layer 220 may be
shorter or smaller than the diameter or area of the lowermost layer
of the second light condensing layer 230 in order to make the depth
detection pixels 120 have a longer focal length than the focal
length of the color detection pixels 110. To be specific, the
diameter or the area of the first layer 221 of the first light
condensing layer 220 may be shorter or smaller than the diameter or
area of the third layer 231 of the second light condensing layer
230. Also, when the thickness T1 of the first light condensing
layer 220 is the same as the thickness T2 of the second light
condensing layer 230, the lowermost layer of the second light
condensing layer 230 may be thinner than the lowermost layer of the
first light condensing layer 220. To be specific, when the
thickness T1 of the first light condensing layer 220 is the same as
the thickness T2 of the second light condensing layer 230, the
thickness t3 of the third layer 231 of the second light condensing
layer 230 may be thinner than the thickness t1 of the first layer
221 of the first light condensing layer 220.
[0062] Also, referring to FIGS. 3A, 4A and 5, the pixels of the
pixel array of the image sensor in accordance with the first or
second embodiments may further include a plurality of phase
difference detection pixels 130. The phase difference detection
pixels 130 may include a first pixel 131 and a second pixel 132
that respectively include a light blocking layer 251 having an
opening 251A and a light blocking layer 252 having an opening 252A.
The openings 251A and 252A are eccentrically disposed in different
directions with respect to an optical axis (or central axis) of the
photoelectric conversion layer 202.
[0063] The phase difference detection pixels 130 may include the
photoelectric conversion layer 202, the dielectric layer 205
disposed over the photoelectric conversion layer 202, the light
blocking layers 251 and 252 formed inside the dielectric layer 205
and having the opening 251A and the opening 252A that are
eccentrically disposed in different directions with respect to the
optical axis of the photoelectric conversion layer 202, and a third
light condensing layer 240 disposed over the dielectric layer 205.
The third light condensing layer 240 may be a hemispheric lens or a
stacked structure where a plurality of material layers is stacked.
An upper layer of the plurality of material layers has a shorter
diameter than a lower layer thereof. The third light condensing
layer 240 may be the same as the second light condensing layer 230
in material.
[0064] As described above, the image sensor in accordance with the
first or the second embodiment may suppress interference from
external environments (e.g., high illumination environment) and
generate more accurate depth data by making the color detection
pixels 110 and the depth detection pixels 120 have different focal
lengths. Also, since the image sensor has a simple structure, it
may be easily mass-produced resulting in lower production
costs.
[0065] Additionally, by employing the phase difference detection
pixels 130, the image sensor may generate the depth data more
accurately.
[0066] Hereafter, a method for generating depth data using the
image sensor, according to an embodiment, will be explained by
referring to FIGS. 6 and 7.
[0067] FIG. 6 is a block diagram illustrating an Image Signal
Processor (ISP) of an image sensor in accordance with an
embodiment. FIG. 7 is a flowchart describing a method for
generating depth data using an image sensor in accordance with an
embodiment.
[0068] Referring to FIGS. 1 to 6, the image signal processor 1110
receives a digital pixel signal (DPS) outputted from the pixel
array through the Correlated Double Sampler (CDS) 1130 and the
Analog-Digital Converter (ADC) 1140 and processes the digital pixel
signal. The image signal processor 1110 may include a control block
310, a memory block 320, a phase difference detection block 330
including a blur weight value generation unit 331, an edge
detection block 340, and a depth detection block 350.
[0069] The control block 310 may perform a function of controlling
operations of the image signal processor 1110. The memory block may
perform a function of storing the digital pixel signal transmitted
to the image signal processor 1110 and the information processed in
the image signal processor 1110. The phase difference detection
block 330 may perform a function of measuring the focal length and
generating Depth Of Focus (DOF) information based on the output
signals of the phase difference detection pixels inside the pixel
array or signals provided by a phase difference detection sensor.
The blur weight value generation unit 331 of the phase difference
detection block 330 may generate a blur weight value based on the
focal length that is needed to generate accurate depth data. The
edge detection block 340 may perform a function of extracting an
edge value from the digital pixel signals of the color detection
pixels and the depth detection pixels. The depth detection block
350 may perform a function of generating depth data based on the
data provided by the edge detection block 340 and the phase
difference detection block 330 including the blur weight value
generation unit 331.
[0070] Hereafter, a method for generating depth data by using an
image sensor in accordance with an embodiment will be described
with reference to FIGS. 6 and 7.
[0071] First, in step S101, the memory block 320 loads with Digital
Pixel Signals (DPS) of a predetermined group. The Digital Pixel
Signals (DPS) of the predetermined group refer to a minimum amount
of data necessary to generate the depth data. For example, when
digital pixel signals are provided row by row, the digital pixel
signals of the predetermined group may be data included in a single
row or may be a sum of data included in several to tens of
rows.
[0072] The digital pixel signals of the predetermined group may
include output signals of the color detection pixels, output
signals of the depth detection pixels, and output signals of the
phase difference detection pixels.
[0073] Subsequently, in step S102, the focal length from the
digital pixel signals of the predetermined group loaded on the
memory block 320 to the object is measured. The focal length to the
object may be calculated in the phase difference detection block
330 based on the output signals of the phase difference detection
pixels, and it may also be calculated from a deviation amount based
on a phase difference.
[0074] Subsequently, in step S103, it is decided whether the
measured focal length coincides with the focal length set by a
user. The measurement of the focal length performed in the step
S102 is repeatedly performed until the measured focal length
coincides with the focal length set by the user.
[0075] Subsequently, in step S104, the blur weight value is
calculated. The blur weight value may be calculated by generating
Depth Of Focus (DOF) information in the phase difference detection
block 330 and sampling a phase difference in the blur weight value
generation unit 331 using the generated DOF information as a
reference value. The phase difference refers to a difference
between a phase at a front-focusing (or in-focus) pin where a focus
is made in front of the object and a phase at a back-focusing (or
out-focus) pin where a focus is made behind the object.
[0076] Subsequently, in step S105, a first edge value is extracted
using the output signals of the color detection pixels. The output
signals of the color detection pixels may be the signals of the
first color pixels to the third color pixels, and the first edge
value may be an edge value of color data generated in the color
detection pixels. For example, the first edge value may be an edge
value of RGB data generated in the red pixels, the green pixels,
and the blue pixels.
[0077] Subsequently, in step S106, whether the first edge value is
extracted or not is decided. The first edge value is an essential'
parameter for generating the depth data. Therefore, if the first
edge value is not extracted, the memory block 320 loads with
digital pixel' signals of the next group in step S201 and the
above-described processes are repeated.
[0078] Subsequently, in step S107, a second edge value is extracted
by using the output signals of the color detection pixels. To
extract the second edge value, some of the output signals of the
color detection pixels and the output signals of the depth
detection pixels may be used. To be specific, the second edge value
may be extracted by using the depth detection pixel signals
generated from infrared rays and the color detection pixel signals
generated from the red pixels (which may be the first color pixels)
of a wavelength band adjacent to the infrared rays.
[0079] Subsequently, in step S108, whether the second edge value is
extracted or not is decided. The second edge value is an essential
parameter for generating the depth data along with the first edge
value. Therefore, if the second edge value is not extracted, the
memory block 320 loads with digital pixel signals of the next group
in step S201 and the above-described processes are repeated.
[0080] Subsequently, in step S109, a first blur value and a second
blur value are generated. The first blur value and the second blur
value may be generated by calculating size values of blurs of the
first edge value and the second edge value that are extracted based
on a Point Spread Function (PSF), and reflecting the blur weight
value calculated in advance into the calculated size values of the
blurs, respectively. The size values of the blurs calculated based
on the first and second edge values and the point spread function
may be amplified by reflecting the blur weight value.
[0081] Subsequently, in step S110, depth information is generated
based on a difference between the first blur value and the second
blur value.
[0082] According to an embodiment, an image sensor includes color
detection pixels and depth detection pixels with different focal
lengths that are integrated in a pixel array to suppress
interference from the external environment, such as a high
illumination environment, and generate accurate depth data. Also,
with a simple structure, the image sensor may be easily
mass-produced with a lower production cost.
[0083] In addition, the phase difference detection pixels may
further improve accuracy of the depth data of the image sensor.
* * * * *